Feature/sg 1386 granular control over export in ptq and qat #1879
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ofrimasad
reviewed
Feb 29, 2024
…rt-in-ptq-and-qat
…rt-in-ptq-and-qat
ofrimasad
reviewed
Mar 5, 2024
…-over-export-in-ptq-and-qat' into feature/SG-1386-granular-control-over-export-in-ptq-and-qat
BloodAxe
deleted the
feature/SG-1386-granular-control-over-export-in-ptq-and-qat
branch
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR introduce a better control on the model export to ONNX format while using PTQ and QAT.
It introduce an
ExportParamsdataclass andexport_paramsargument toptq()andqat()methods that is meant to contain many properties one may use to explicitly set this or another parameter used during export process.As an example one may enable of disable attaching postprocessing to detection model. And this would behave exactly the sample way for PTQ and QAT (Currently these methods export model differently #1189)
Motivation to introduce an
ExportParamsdataclass is to introduce some form of strict schema of what parameters can be passed. This is especially useful in Colab environment since once can easily get code completion and navigate to that dataclass and read comments on each field.A recipe-based entry-points (Like
quantize_from_config) can easily populate this object as it has no mandatory fields.To PR should be merged after #1860, so that we can incorporate .export() support for segmentation models.